Latest Articles for Fine-tuning

Exploring the role of machine learning in predicting material behaviors and challenges faced.

A study on aligning agents in 3D games to improve behavior.

Learn how to train models for text embeddings wisely and effectively.

UltraMedical collections improve medical language models and address data shortages.

Discover how LoCalPFN improves transformer performance on tabular data.

Study reveals effective techniques to enhance multimodal large language models.

A study on the effectiveness of various lightweight models in image classification.

This study explores methods to enhance vision-language models using generated images.

This article reviews methods to enhance dialogue generation in language models.

Examining the risks and safety measures in fine-tuning language models.

A look into how LLMs tackle programming by example challenges.

A new approach to classifying tabular data using ICL-transformers shows promising results.

Examining the effectiveness of reasoning in large language models.

Investigating how latent space affects transformer model performance on language tasks.

Examining the impact of synthetic news content and detection difficulties.

Examining memorization in code completion models and its privacy implications.

This article examines ways to improve planning abilities in large language models.

A method to evaluate model knowledge through internal processing.

DetectBench evaluates LLMs on their ability to detect hidden evidence in reasoning tasks.

A new method to improve model stability and performance in low-resource settings.

How fine-tuning affects language models' ability to recall facts accurately.

Advancements in fine-tuning language models using innovative techniques.

RankAdaptor optimizes fine-tuning for pruned AI models, enhancing performance efficiently.

Methods to reduce memory usage during fine-tuning of large models.

This study presents a dataset and method to enhance Chinese ASR accuracy using Pinyin.

New methods refine reasoning skills in language models for better task performance.

A new method enhances how language models align with human values.

This study focuses on enhancing model responses by targeting specific length requirements.

Research on improving knowledge transfer in resource-limited smart devices.

This study evaluates how well large language models use external information.

GTZAN-synth dataset leverages synthetic music for better music tagging systems.

New method enhances spiking neural networks' performance in language tasks.

New methods improve molecular design by measuring prediction uncertainty.

A new system enhances data processing while ensuring user privacy and efficient resource use.

HyperLoader improves multi-task model training using innovative techniques and hypernetworks.

Research shows how easily safety features can be removed from Llama 3 models.

A new framework enhances large model performance efficiently during fine-tuning.

CPT improves black-box model performance without direct access to internal parameters.

Fine-tuning large language models directly on smartphones while protecting user data.

Examining methods to enhance code generation for specialized programming languages using LLMs.